Through our research we aim to develop quantitative methods for the assessment of motor learning in stroke survivors using the lower extremity rehabilitation Lokomat platform (Hocoma AG, Switzerland). Accurate understanding of how the structure of a rehabilitation program affects the long-term retention of motor skills is essential for establishing the efficacy of therapeutic approaches and for comparing efficacy across strategies. Current methods are also very laborious for rehabilitation workers and in some situations strain on the therapist can be the limiting factor on the length of a session or on the number of activities conducted per session. By applying advanced engineering techniques from the fields of control theory and signal processing, we hope to develop automated and adaptive strategies for devices like the Lokomat platform that allow rehabilitation workers to provide their skills while removing much of the strain. These strategies may result in increased speed of recovery by providing patients with tasks dictated by their individual baseline and rate of improvement without the need for therapist intervention.